Policy Information Report

Locked Up and Locked Out:
An Educational Perspective on
the U.S. Prison Population

Listening.
Learning.
Leading.

This report was written by:
Richard J. Coley
Paul E. Barton
Educational Testing Service
The views expressed in this report
are those of the authors and do not
necessarily reflect the views of the
officers and trustees of Educational
Testing Service.
Additional copies of this report can
be ordered for $15 (prepaid) from:
Policy Information Center
Mail Stop 19-R
Educational Testing Service
Rosedale Road
Princeton, NJ 08541-0001
(609) 734-5949
pic@ets.org
Copies can be downloaded from:
www.ets.org/research/pic

Table of Contents
Preface .......................................................................................................... 2
Acknowledgments ........................................................................................ 2
Highlights ..................................................................................................... 3
Introduction ................................................................................................. 5
The Exploding Prison Population .............................................................. 7
The Juvenile Detention System................................................................. 11
The Prison Education Enterprise ............................................................. 13
What Prison Education and Training Programs Accomplish................. 19
Prisoner Re-Entry Programs: A Recent Response................................... 22
The Children of Prisoners ......................................................................... 24
The Special Case of the Black Male.......................................................... 27
Conclusion.................................................................................................. 30

Copyright © 2006 by
Educational Testing Service.
All rights reserved. Educational
Testing Service is an Affirmative
Action/Equal Opportunity
Employer. Educational Testing
Service, ETS, and the ETS logo
are registered trademarks of
Educational Testing Service.

February 2006
Policy Evaluation and
Research Center
Policy Information Center
Educational Testing Service

1

Preface
In this important and sobering report, Richard J. Coley
and Paul E. Barton provide a broad perspective on the
U.S. prison population and offer judgments about the
status of prison education programs.
The prison education enterprise is perhaps more
important now than ever, as the prison population
surges and evidence accumulates about the effectiveness of prison education programs on recidivism. Yet
this population continues to be under-educated, with
most prisoners having less than a high school education. Ever-larger numbers of ex-prisoners are returning
to their communities poorly prepared to re-enter the
workforce and, as a result, to support themselves and
their families, or to form families and rear children.
It is a formula for disaster — for more crime, more
recidivism, and greater cost to society.

In the face of this demonstrated need, we should be
alarmed that we are losing ground in the prison education enterprise; investment in correctional education
programs is not keeping pace with the exploding population of prisoners. Coley and Barton issue a challenge
for society to support this important investment and to
acknowledge the plight of prisoners’ children — children
whose chances of following in their parents’ footsteps
are high, unless we have the will to break the cycle.

Michael T. Nettles
Vice President
Policy Evaluation and
Research Center

Acknowledgments
The authors wish to thank the following individuals for
their help with this report. The report was reviewed
by John Linton, Office of Safe and Drug-Free Schools,
U.S. Department of Education; Caroline Wolf Harlow,
Bureau of Justice Statistics, U.S. Department of Justice; and Stephen J. Steurer, Executive Director, Correctional Education Association. In addition to providing

2

feedback on the report, the reviewers provided valuable
guidance on the data that are available to address this
topic. The report was edited by Janet Levy and proofread by Alex Dering. Marita Gray was the photographer
and designer for the cover. Christina Guzikowski provided layout. Errors of fact or interpretation are those
of the authors.

Highlights

This report brings together data and information from
a variety of sources to provide an educational perspective on the nation’s prison population. It examines
the size and nature of that population and provides
information on trends when possible. The report also
describes the limited information that is available on
the prison education enterprise and summarizes what
is known about the effects of education and training
on recidivism. Finally, it provides a perspective on the
children whose development and lives are negatively
affected by incarceration and on the plight of young
Black males, who are increasingly coming into contact
with the correctional system.
Some highlights of the report are provided below:
The size of the prison population continues to explode, even as the crime rate shrinks.
The rate of incarceration has surged, more
than doubling from 313 per 100,000 people
in 1985 to 726 in 2004. Over the same period,
victimizations data show a fairly dramatic and
steady decrease in crime. A variety of explanations are given for this seeming paradox.
Prisons bulge with poorly educated inmates, and
as this population grows, the related investment in
education and training is not keeping pace.
This low and declining investment contrasts
with an increasing body of research showing
that education and training programs can raise
employment prospects and cut recidivism.
The public is realizing that bulging prisons also
mean that large numbers of ex-prisoners will
return to their communities with three strikes
against them for getting a job — an essential step
to going straight.
Longer sentences or not, most prisoners come
back to the community. Hindered by the following barriers, ex-prisoners are less likely to
become self-supporting — and therefore, less
likely to succeed in society:
• Strike One — Ex-inmates with little education and low literacy levels are not desired by
employers.

• Strike Two — Employers are looking for
employees who have had steady and successful
work experiences, even for low-skilled jobs. Exprisoners disproportionately don’t have them.
• Strike Three — Many jobs are “off limits” to
ex-prisoners.
Some prisons place soon-to-be-released prisoners in short-term “prisoner re-entry” programs.
While such programs are welcome and may be
effective, there is a need to buttress them with
solid, longer-term programs.
There is minimal state data on prisoner education regarding enrollments, completions, degrees
received, test scores, etc.
Data available for 1993-94 show huge disparities among the states in many dimensions of
the correctional education enterprise. Federal
and state justice, corrections, and education
departments must collaborate to get the data
needed to judge the reach and effectiveness of
prison education and training programs.
While punishment is wholly appropriate for criminals, it is not appropriate for the more than 1.5
million children of prisoners, who are most disserved by the corrections system.
Neglecting these unintended victims will likely
lead to these children replacing their parents in
the prisons of the future.
The incarceration of young Black males — particularly high school dropouts — has reached levels
that jeopardize the achievement of broader socialjustice goals.
The incarceration rate for Black 25- to 29year-old males, for example, is 13 percent,
compared with 2 percent of the White and 4
percent of the Hispanic populations in that age
group. For young Black males without a high
school diploma, about as many are in prison as
are employed.
It is estimated that more than half of all Black
males who do not have a high school diploma
have a prison record, compared to one in 10

3

White males. The dire employment prospects
of Black male dropouts affect the likelihood for
success in marriage, child rearing, and ensuring that the next generation helps to close the
achievement gap.
While in prison, inmates have time to obtain
their high school diplomas, train for a job,
and prepare to earn a living when they return
to their communities. Prisons should provide
such opportunities and push for prisoners to
take advantage of them.
While this country has not ratcheted up its investments in correctional education while adopting a
“get tough on crime” approach, it must recognize
that providing prisoners with the education and
job skills they need to stay out of prison can save
scarce resources in the long run.
The “chain gang” was tough physically; the
“learning gang” requires hard mental effort
and discipline. The public suffered when the
prisoner’s original crime was committed;
the potential for damage increases when the
prisoner returns to society without a means of
making it in the employment world.

4

Introduction

The first true American prison was the Walnut Street
Jail in Philadelphia, created by the Quakers in 1791.
The prison had three objectives: to ensure public security, reformation of prisoners, and “humanity toward
those unhappy members of society.” Elaborating on the
last two objectives, the jail’s inspectors reported that
“edifying persons have at all times access to the prisoners.” Furthermore, the architects added a school to the
prison in 1798 as “the most beneficial…for learning for
some and improving for others in the first principles of
reading, writing, and arithmetic.”1

prospered. When rehabilitation has been out of favor,
prison education has languished. The use of education
and training in prisons spread in the 1930s, receded,
and then came back into favor in the 1960s. Since the
1980s, “tough treatment” has been the trend. According to Schlossman and Spillane, “Correctional education was largely excluded from the main currents of
prison reform during the 1980s,” when opinion polls
showed that Americans became “increasingly hostile
and suspect of all rehabilitative programs aimed at
reintegrating prisoners into the mainstream.”4

Since then, education has grown throughout our
prison system — as has the controversy over rehabilitation versus punishment. As early as the 1820s,
Samuel Hopkins of the New York Legislature was arguing that “inmate life had not been sufficiently severe
and should produce more terror and suffering.” Such
views gave rise to the Auburn, N.Y., system, which
subscribed to the belief that “too much faith had been
placed in [the convict’s] reformability.” Thus, education got little attention in the Auburn system.2

During the 1980s and 1990s, state and federal
spending for correctional education programs
decreased significantly. Today, because of state budget
problems, states such as California, Florida, and
Illinois are cutting correctional education budgets
even further.5 At the federal level, Congress passed a
law in 1994 that prohibited inmates from receiving
Pell Grants, effectively defunding postsecondary
education in prisons.

During the late nineteenth century, Superintendent
of the New York State Reformatory at Elmira, Zebulon
Brockway, became known across the nation for his use
of education and training in prison. A report by Steven
Schlossman and Joseph Spillane states that Brock
way “placed the academic programs (and later the
vocational programs) into the hands of professional,
full-time teachers who were drawn from the commu
nity.” Sentences were indeterminate, and time served
became heavily dependent on participation and performance in the education and training system.3
Especially noteworthy for the purpose of this
report is that, when the rehabilitation approach to
corrections has been in favor, prison education has

These lean times for prison education programs
coincide with an explosive surge in the size of the
nation’s prison population, creating somewhat of a
“perfect storm.” While the danger is clear, many
obstacles — both financial and attitudinal — block
the path to a safer harbor.
Commenting on basic economics about crime and
education, Stephen Steurer, executive director of the
Correctional Education Association, notes: “Crime is
not partisan. Felons are opportunistic. They attack
Republicans and Democrats indiscriminately. All of us
want to be safe and secure. Public policy on crime and
punishment should be determined by the most effective crime prevention and reduction technique available through proven research.”6

For this early history, see Negley K. Teeters, The Cradle of the Penitentiary: The Walnut Street Jail at Philadelphia, 1773-1835, sponsored by
the Pennsylvania Prison Society, 1955.
2
Walter Silva, “A Brief History of Prison Higher Education in the United States,” in M. Williford (ed.), Higher Education in Prison: A Contradiction in Terms?” Phoenix, AZ: Oryx Press, 1994.
3
Steven Schlossman and Joseph Spillane, Bright Hopes, Dim Realities: Vocational Innovation in American Correctional Education, National
Center for Research in Vocational Education, University of California, Berkeley, March 1992.
4
Schlossman and Spillane, 1992.
5
Audrey Bazos and Jessica Hausman, Correctional Education as a Crime Control Program, UCLA School of Public Policy and Social
Research, Department of Policy Studies, March 2004.
6
Stephen Steurer, Executive Director, Correctional Education Association, remarks prepared for the Detroit Free Press.
1

5

The growing prison population is described in the
first section of this report and is followed by a description of the juvenile detention system. The third section, on the prison education enterprise, discusses the
limited information that is available on the size and
characteristics of correctional education programs
across the states. The accumulating evidence on the
effectiveness of correctional education programs is
described next. Re-entry programs, increasing in
response to the surging outflow of prisoners to society,
are described in the following section. Often neglected
in the policy debate, the effects of parental incarceration on children are described next; this section
is followed by a discussion on the predominance of
Black males in correctional institutions. Some overall
conclusions are drawn in the last section.

6

The Exploding Prison Population

This section provides some overall context for the topics
discussed in this report by describing the size of the
prison population in the United States and how it has
changed over time, both in absolute and in relative
terms. The prison population is then profiled in terms
of its racial/ethnic composition, its age distribution, its
educational attainment, and its literacy level.

Figure 2
Number of Persons Held in State or Federal
Prisons or in Local Jails, 1985 to 2004
2.5

Total

2,131,180

2

There are a number of statistics that can be viewed
when trying to size up the nation’s correctional system. When we total the adults who, in 2003, were on
probation, in jail or prison, or on parole, the number
approaches 6.9 million. This represents an overall
increase of 274 percent since 1980. The breakdown is
shown in Figure 1, along with specific percent changes
since 1980.

Figure 1

1.5

State

1,241,034

1

Local

713,990

0.5

Federal
0

85

169,370

90 91 92 93 94 95 96 97 98 99 00 01 02 03 04

Source: U.S. Department of Justice, Sourcebook of Criminal Justice Online (http://www.
albany.edu/sourcebook/pdf/t612.pdf) and Harrison and Beck, 2005.

Number of Adults on Probation, in Jail or Prison,
and on Parole, 2003
Probation

Jail

Prison

Parole
6.9 million

4,073,987

691,301 1,387,269 774,588

Percentage Change, 1980 to 2003
Probation +264 percent
Jail
+279 percent

Prison +334 percent
Parole +251 percent

Source: U.S. Department of Justice, Sourcebook of Criminal Justice Online (http://www.
albany.edu/sourcebook/pdf/t61.pdf).

Another statistic, and one on which we’ll focus in
this discussion, is the number of persons who are
incarcerated in federal or state prisons or local jails.
At midyear 2004, the nation’s prisons and jails held
2,131,180 persons. Of those, 7.9 percent were in federal custody; 58.2 percent were in state custody; and
33.5 percent were held in local jails.7 Figure 2 shows
the trends since 1985.

7

Millions

The Size of the Population

Since 1985, the number of people incarcerated
has jumped from about 744,000 to almost 1.6 million
in 1995, to more than 2.1 million in 2004. That
represents an overall increase of 186 percent. While all
sectors have grown over that time period, the highest
growth was in the federal prison population, which
increased by 373 percent. Increases in the other sectors ranged from 175 percent in state prisons to 178
percent in local jails.
While the tremendous growth in the absolute number of individuals who are incarcerated is itself of concern, we also need to examine the relative growth of
that population. Figure 3 shows trend data on the rate
of persons incarcerated per 100,000 U.S. residents.
Between 1985 and 2004, the rate has soared from 313
to 726.
Given those data one might reasonably expect to see
an increase in crime over those years, increases that
would reflect why more criminals were incarcerated.
That is not the case, however.

Paige M. Harrison and Allen J. Beck, Prison and Jail Inmates at Midyear 2004, U.S. Department of Justice, Bureau of Justice Statistics
Bulletin, April 2005.

7

Figure 3
Rate (per 100,000 Residents) of Persons Held in
State or Federal Prisons or in Local Jails, 1985 to
2004
Persons in Custody per 100,000 Residents

800
726
700
600
500
400

Truth-in-sentencing laws have also prolonged the
amount of time inmates remain incarcerated by
requiring that prisoners serve a substantial portion
of their sentences. These laws — intended to reduce or eliminate early release for good behavior or
due to prison overcrowding — have dramatically increased the time inmates must serve. Other factors,
including increasing numbers of parolees returned
to prison for technical violations and the difficult
economic conditions in the early 1990s, may also
have contributed to rising incarceration rates.” 9

313

Characteristics of the Prison Population

300
200
85

90 91 92 93 94 95 96 97 98 99 00 01 02 03 04

Sources: U.S. Department of Justice, Sourcebook of Criminal Justice Statistics Online
(http://www.albany.edu/sourcebook/pdf/t612.pdf) and Harrison and Beck, 2005.

An examination of various crime reports reveals a
fairly dramatic and steady decrease in crime over the
past 30 years. One example is the number of total violent crimes (NCVS) reported by the Bureau of Justice
Statistics. According to that measure, there has been
a consistent decline from 3.6 million violent crimes in
1973 to 1.8 million in 2003.8
Then what are the reasons for rising incarceration
rates? The U.S. Department of Education’s Office of
Safe and Drug-Free Schools offers the following
explanation:
“The unprecedented growth in prison populations
can be traced, in part, to new federal and state sentencing guidelines that have imposed mandatory
prison terms and lengthened minimum sentences
for repeat offenders. In particular, the introduction
of ‘Three Strikes’ legislation — enacted federally in
1994 and implemented by several states during the
decade — has increased the time inmates remain
incarcerated. While the specifics vary across states,
individuals committing a second or third offense
may face double or triple the prison sentence they
would otherwise have received.

This section examines the overall age, racial/ethnic
breakdown, and educational level of the prison population. In order to focus this discussion, data are provided for males, since males make up more than 90
percent of this population.10
Incarceration rates by age and race/ethnicity are
shown in Figure 4. For each age group, the figure
shows the percentage of that population in federal or
state prisons or local jails in 2004.
Overall, 1.3 percent of males were incarcerated in
2004. The comparative figures for Black, Hispanic,
and White males were 4.9, 1.7, and 0.7 percent, respectively. The incarceration rates for Black men are
particularly troubling, especially for those in their 20s
and 30s. Among males ages 25 to 29, nearly 13 percent
of Black males were incarcerated, compared to nearly
4 percent of Hispanic males and nearly 2 percent of
White males. Among the more than 2.1 million incarcerated as of June 30, 2004, an estimated 576,600
were Black males between the ages of 20 and 39. And
while incarceration rates drop with increasing age, the
percentage of Black males from ages 45 to 54 in prison
(nearly 5 percent) was more than six times the rate for
White males. The data for female prisoners show similar racial/ethnic differences, although rates are much
lower than for males.11

Bureau of Justice Statistics, http://www.ojp.usdoj.gov/bjs/glance/tables/4meastab.htm.
Steven Klein et al., Correctional Education: Assessing the Status of Prison Programs and Information Needs, U.S. Department of Education,
Office of Safe and Drug-Free Schools and MPR Associates, 2004.
10
According to the Bureau of Justice Statistics, 183,400 females were incarcerated in 2004 compared to 1,947,800 males.
11
Harrison and Beck, 2005.
8
9

8

Educational Level of Prisoners
Surveys of the characteristics of prisoners are few and
far between. The most recent survey, in 1997, found
that about 41 percent of the nation’s incarcerated had
less than a high school education. Subgroups of state
prison inmates who had not completed high school or
the GED include:

Figure 4
Percentage of the Male Population in State or
Federal Prisons or Local Jails, by Race/Ethnicity
and Age, June 30, 2004

Age
All

18-19

20-24

25-29

30-34

Total
Black
Hispanic
White

45-54

55 or older

5.4

Figure 5

2
0.9

11.1
3.6

8th grade or less

1.6

White
Black
Hispanic

3.4
12.6

11
12
28

3.6

Some high school

1.7

White
Black
Hispanic

3.1
11
3.4
1.7

2.9
2.8

35
25
25

High school diploma

2.2

White
Black
Hispanic

8
2.4
1.3

23
21
15

Some postsecondary

1.2

White
Black
Hispanic

4.6
1.7
0.7

Total
Black
Hispanic
White

32
25

White
Black
Hispanic

10

Total
Black
Hispanic
White

16

GED
2.8

11
8
6

College graduate or more

0.3
0.9
0.5

White
Black
Hispanic

0.2

0

4

8

Percentage
Source: Calculated from data in Harrison and Beck, 2005.

12

Educational Attainment of State Prison Inmates,
by Racial/Ethnic Group, 1997

3.3

Total
Black
Hispanic
White

40-44

• 52 percent of inmates age 24 or younger12

1.7

Total
Black
Hispanic
White

Total
Black
Hispanic
White

• 27 percent of Whites

0.7

Total
Black
Hispanic
White

35-39

• 44 percent of Blacks

4.9
1.7

Total
Black
Hispanic
White

Total
Black
Hispanic
White

• 53 percent of Hispanics

1.3

12

16

4
2
2

0

10

20

30

40

Percentage
Source: Caroline Wolf Harlow, Education and Correctional Populations, U.S. Department of
Justice, Bureau of Justice Statistics Special Report, January 2003.

Caroline Wolf Harlow, Education and Correctional Populations, U.S. Department of Justice, Bureau of Justice Statistics Special Report,
January 2003.

9

Literacy Levels of Prisoners
Given these educational attainment levels, it is not surprising that the literacy levels of prisoners are low as
well. Literacy data collected on the prison population
as part of the National Adult Literacy Survey (NALS)
provide prose, document, and quantitative literacy
profiles for this population. On each scale, the literacy
level of prisoners is substantially lower than that of the
U.S. population as a whole.
Figure 6 compares the average literacy scores of
prisoners with average scores for the total population
for each literacy scale. While these data are for 1992, the
educational profile of the prison population has changed
little, making it unlikely that the literacy levels have improved. New adult literacy data will be available in 2006.

Moreover, prisoners were far more likely than the
national population to perform in the lowest literacy
levels on each of the three scales and far less likely to
attain the highest levels. As shown in Figure 7, about
one-third of prisoners scored in the lowest levels of
prose, document, and quantitative literacy, and another
third performed in the second-lowest level.13

Figure 7
Percentage of Prisoners and Total U.S. Population Who
Performed in Each Proficiency Level, by Literacy Scale
50

20

26

21
17

6
1

2

3

Low

40

Percentage

5

High

38

30

31

28

25

23
20

15

10
4

Document Literacy

1

50

240

40

Quantitative Literacy
271

Total Population

Percentage

40

Prisoners

2

3

Low

267

Prisoners

High

Quantitative Literacy
31

25

22

22

20

17

236
240
250
260
270
Average Proficiency Score (0 to 500)

Source: National Adult Literacy Survey, 1992.
13

280

0

0
5

Proficiency Level

32
30

3

4

10

230

0

4

Document Literacy

0

Total Population

3

Proficiency Level

33

246

Prisoners

27

0

Average Literacy Proficiencies of Adults in the
Total and Prison Populations, by Literacy Scale
273

32

31

10

50

Total Population

Total U.S. Population
Prison Population

37

30

Figure 6

Prose Literacy

Prose Literacy

40

Percentage

Figure 5 shows the educational attainment levels for
state prison inmates in 1997, broken out by racial/ethnic
group. Overall, Black and Hispanic state prison inmates
had much lower levels of educational attainment than
White inmates: 53 percent of Hispanic and 44 percent
of Black inmates had not graduated from high school or
earned a GED, compared to 27 percent of White inmates.

6
1

Low

2

3

Proficiency Level

4

4

1
5

High

Source: National Adult Literacy Survey, 1992.

For more detail on the literacy of the prison population, see Paul E. Barton and Richard J. Coley, Captive Students: Education and Training
in America’s Prisons, Policy Information Report, Policy Information Center, Educational Testing Service, January 1996; and Karl O. Haigler
et al., Literacy Behind Prison Walls: Profiles of the Prison Population from the National Adult Literacy Survey, prepared by Educational Testing Service under contract with the National Center for Education Statistics, U.S. Department of Education, October 1994.

10

The Juvenile Detention System

Early in U.S. history, children who broke the law were
treated the same as adult criminals. But by the turn of
the twentieth century, 32 states had established juvenile
courts. Rather than merely punishing delinquents for
their crimes, juvenile courts sought to turn delinquents
into productive citizens through treatment. Through
the 1950s, most juvenile courts had exclusive original jurisdiction over all those under age 18 who were
charged with a crime; only if that court waived its jurisdiction could a child be transferred to a criminal court.
In 1968, Congress passed the Juvenile Delinquency
Prevention and Control Act, which recommended
that children charged with non-criminal offenses be
handled outside the court system. In 1974, Congress
passed the Juvenile Justice and Delinquency Prevention Act. This act required states receiving federal
grants to reform their systems, and addressed violent
crime by focusing on prevention, intervention, and
accountability. Over the past two decades, there has
been a shift toward treating more juvenile offenders as
criminals. States have been trying to balance system
and offender accountability, offender competency development, and community protection.14
Juvenile crime statistics indicate that much progress has been made. For example, in 1974, arrests of
juveniles for violent crimes had increased 216 percent
from 1960. The statistics continued to be troubling
into the early 1990s, with substantial growth in the
number of juveniles arrested peaking in 1994, when
there were 2.7 million juvenile arrests. By 2002, the
number of arrests had dropped to 2.3 million, and
the percentage of all violent crimes committed by
juveniles was down considerably from 1974. Finally,
between 1994 and 2002, the juvenile arrest rate for
violent crimes fell to its lowest level since at least 1980;
and between 1993 and 2000, the juvenile arrest rate for
murder fell 72 percent.15
While this progress is encouraging, many youth remain in custody in their respective states. The United
States does not have a juvenile justice system, but 51

separate systems. While describing these state systems
is beyond the scope of this report, interested readers
can find profiles of state systems via the National Center
for Juvenile Justice (http://www.ncjj.org/stateprofiles).
Across all the states in 2001, there was a total of
104,413 juveniles in residential placement. Of these,
85 percent were male and half were between 16 and
17 years old. The resulting detention rate of 336 per
100,000 juveniles represents a decline from 361 in
1999 and 359 in 1997. Figure 8 shows the rate for
2001, broken out by race/ethnicity.

Figure 8
Rate per 100,000 of Juveniles in Residential
Custody, by Race/Ethnicity, 2001
863

Black
608

American Indian
Hispanic

366

All groups

336
209

White
120

Asian
0

200
400
600
800
Number per 100,000 of Population

1,000

Source: Melissa Sickmund, T.J. Sladky, and Wei Kang, Census of Juveniles in Residential
Placement Databook, 2004. Online. Available: http://www.ojjdp.ncjrs.org/ojstatbb/cjrp/.

Black juveniles were more than four times as likely
as White juveniles to be in custody, and more than
seven times as likely as Asian juveniles to be in custody. Rates for American Indian and Hispanic juveniles
were also higher than the rate for all groups combined.
Figure 9 shows the most serious juvenile offenses
for 2001, broken into several categories.16 Ninety-five
percent of juveniles in residential placement were
there because of delinquency; the remaining 5 percent

U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention, Juvenile Justice: A Century
of Change, Juvenile Justice Bulletin, December 1999.
15
Federal Advisory Committee on Juvenile Justice, Annual Report 2004. Recommendations Report to the President and the Congress of the
United States, January 2005.
16
The categories are defined as follows: “Delinquency” is an offense that is also considered illegal for adults. “Status” is a non-delinquent/
non-criminal offense; an offense that is illegal for underage persons but not adults. “Person,” “property,” and “public order” are offenses
against persons, property, and public order, respectively.
14

11

were there for “status” offenses, such as being truant
or running away. Twenty-three percent of juveniles in
residential placement had committed violent crimes,
and another 23 percent had committed property
crimes, such as stealing or vandalizing. Girls were
more likely than boys to be in residential placement
because of status offenses (14 percent versus 3 percent).17

To help put juveniles on a path to a crime-free life,
logic dictates that residential facilities should also
function as good schools — from both academic and
vocational standpoints.

Figure 9
Most Serious Juvenile Offenses, 2001

Delinquency
Person Offenses
8,520
7,999
7,328
6,779

Aggravated assault
Simple assault
Robbery
Sexual assault
Other
Criminal homicide

3,220
1,069

Property Offenses
11,435

Burglary
Auto theft
Theft
Other
Arson

6,131
5,856
4,906
1,045

Drug Offenses
Other
Trafficking

6,654
2,432

Public Order Offenses
6,933

Other
Weapons
Alcohol

3,205
313

Status
Incorrigibility
1,983
1,104
Running away
Truancy
784
614
Other
Drinking
519
Curfew 112
0

2 ,000

4,000

6,000

8,000

10,000

12,000

Number of Most Serious Offenses

Source: Melissa Sickmund, T.J. Sladky, and Wei Kang, Census of Juveniles in Residential
Placement Databook. Online. Available: http://ojjdp.ncjrs.org/ojstatbb/cjrp/.

17

Melissa Sickmund, T.J. Sladky, and Wei Kang, Census of Juveniles in Residential Placement Databook, 2004. Online. Available:
http://ojjdp.ncjrs.org/ojstabb/cjrp/. Cited in Child Trends Data Bank, www.childtrendsdatabank.org.

12

The Prison Education Enterprise

The prison education and training enterprise, though
vast in its reach, remains considerably hidden behind
prison walls. Its overall dimensions can be seen from
several perspectives — at least in terms of national
totals and averages. Seldom is there a comprehensive
survey that provides data state-by-state, and the data
that are available do not get down to the level of detail
that allow informed judgments about quality and
effectiveness.

Ideally, this section of the report would detail the
status of educational programs in U.S. prisons and
describe trends in states’ commitments to correctional
education. The data would include information on the
types of education programs available in states, commitments by the states to these programs in terms of
staffing and budgets, levels of inmate participation by
program, and how the prison education enterprise has
changed over time.

The national averages can mask huge differences
among the states in the scope of their education programs. A detailed accounting of what we know, what
we don’t know, and what we need to know is contained
in a recent study, Correctional Education: Assessing the
Status of Prison Programs and Information Needs.18

Unfortunately, the data to support such an effort are
not uniformly available. Our previous report on prison
education benefited from a special data collection by
the Corrections Compendium journal on education programs provided in state correctional systems. The data
we are able to provide in this report are more limited.
Nonetheless, the information we have pulled together
provides a broad overview of what’s happening across

Figure 10
Percentage of Federal, State, and Private Correctional Facilities Providing Education and Counseling
Programs, June 30, 2000

With an education program

89

Secondary
Basic adult education
Vocational training
Special
College
Study release

80
76
54
37
29
11

With a counseling program

96

Drug dependency, counseling, awareness

89

Alcohol dependency, counseling, awareness

88

Life skills, community adjustment

71

Employment

65

Psychological, psychiatric counseling

62

HIV/AIDS

54

Parenting

46

Other

24

Sex offender

20

0

20

40

60

80

100

Percentage of Facilities
Source: James J. Stephan and Jennifer C. Karberg, Census of State and Federal Correctional Facilities, 2000, U.S. Department of Justice, Bureau of Justice Statistics, August 2003.

18

Klein et al., 2004.

13

Table 1
Percentage of Prisons Offering Various Education Programs, 2000
Education Programs Offered

All

Federal

State

Private

Any program

89%

92%

90%

80%

Adult basic

76%

89%

80%

56%

Adult secondary

80%

90%

83%

65%

Vocational training

54%

86%

55%

41%

College coursework

29%

74%

26%

25%

Special education

37%

55%

39%

20%

Study release

11%

6%

8%

27%

Source: James J. Stephan and Jennifer C. Karberg, Census of State and Federal Correctional Facilities, 2000, U.S. Department of
Justice, Bureau of Justice Statistics, August 2003 (revised 10/15/03).

the states. The lack of uniform and comprehensive
data is probably indicative of the health of correctional
education, which is affected by surging prison populations and tight state budgets.
Gathering financial data on the resources spent on
correctional education is difficult for other reasons.
In many states, money for prison education programs
comes out of different agencies — for example, the
state education department, the state department of
corrections, local school districts, local or country
governments, and special districts. Staff costs for these
programs may or may not be allocated to corrections
education budgets. Finally, special arrangements may
also exist whereby local education agencies provide
instruction to inmates.
Availability of Education Programs
Figure 10 shows the percentage of correctional facilities that were providing education and counseling
programs in June 2000. Table 1 shows the percentages separately for federal, state, and private facilities.
According to this survey of correctional facilities, 89
percent of all institutions offered some type of education program — 92 percent of federal, 90 percent of
state, and 80 percent of private facilities. Most of these
institutions provided vocational training (54 percent),
basic adult education (76 percent), and secondary education (80 percent). Special education services, college
classes, and study-release programs were provided by
fewer institutions. Also noteworthy is that 74 percent

19

Klein et al., 2004.

14

of federal prisons offered college coursework, but the
percentage of state and private institutions offering
these programs was much lower.
While most programs increased in number from
1995 to 2000, there was some decline in the number
of federal and state prisons providing adult secondary
education (although there was a substantial increase
in such offerings in private prisons). Vocational training increased in federal and private prisons, but not in
state prisons.19
Finally, the survey breaks out the availability of
education programs by type of facility — confinement
or community based. Confinement facilities were more
likely to provide education programs (94 percent versus 74 percent). This was true for all types of education programs except “study release” programs. These
kinds of data tell us little about the size and scope of
these programs, however.
Availability of Counseling Programs
The survey data shown in Table 2 indicate that 96
percent of institutions provide a variety of counseling
services, most frequently relating to alcohol and drug
dependency and awareness. Prisons also frequently offer programs to help inmates adjust after release. These
programs include life skills and community adjustment
(71 percent) and employment counseling (65 percent).
In general, confinement facilities were more likely than
community-based facilities to provide counseling programs of all types. Again, however, these data provide
no indication of the size or scope of these programs.

Table 2
Percentage of Prisons Offering Various Counseling Programs, 2000
Counseling Programs Offered

All

Federal

State

Private

Any program

96%

92%

97%

92%

Drug dependency, awareness

89%

92%

89%

86%

Alcohol dependency, awareness

88%

92%

88%

85%

Psychological, psychiatric

62%

92%

64%

42%

Employment

65%

88%

62%

70%

Life skills, community adjustment

71%

89%

68%

80%

HIV/AIDS

54%

82%

53%

50%

Parenting

46%

88%

42%

50%

Sex offender

20%

45%

34%

19%

Other

24%

13%

27%

15%

Source: James J. Stephan and Jennifer C. Karberg, Census of State and Federal Correctional Facilities, 2000, U.S. Department of
Justice, Bureau of Justice Statistics, August 2003 (revised 10/15/03).

Table 3
Percentage of Federal and State Inmates Participating in Programs
Since Most Recent Incarceration, 1991 and 1997
Federal Prisons

State Prisons

1991

1997

1991

1997

Total

67%

56%

57%

52%

Adult basic education

10%

2%

5%

3%

GED/high school

27%

23%

27%

23%

Vocational training

29%

31%

31%

32%

College coursework

19%

13%

14%

10%

*

6%

*

1%

8%

6%

3%

3%

English as a second language
Other

* Not Available
Source: James J. Stephan and Jennifer C. Karberg, Census of State and Federal Correctional Facilities, 2000, U.S. Department of
Justice, Bureau of Justice Statistics, August 2003 (revised 10/15/03).

Inmate Participation in Education Programs
The only recent surveys of prisoner participation in
education and training programs are from 1991 and
1997. We do know that investment has been
slipping, so the reach of the programs likely has
declined further.
As shown in Table 3, the overall declines in inmate
participation are substantial in federal prisons, and
less so in state prisons. These declines are pretty much

across the board, except for vocational training, which
went up slightly. Of course, the prison population has
increased, so the actual numbers of inmates in these
programs have increased substantially as well.
According to these data, inmates with the least
amount of education are the most likely to be enrolled
in education programs — around 60 percent compared
with 40 to 50 percent of those with more education.
The participation rates for college graduates are about

15

as high as for those with a high school diploma or with
some college; however, their representation among
the inmate population is, of course, small. Male and
female participation rates are about the same, while
rates for minorities are slightly higher than for White
inmates.
None of this tells us anything about how long
inmates were in classes, how well they did, and what
credentials they were able to achieve. Also, variation
among the states in the size of programs is so huge
that it limits what can be inferred from the national
statistics. This variation was evident in the comprehensive survey published by the Corrections Compendium and summarized in the 1996 ETS report Captive
Students: Education and Training in America’s Prisons.
Eligibility requirements for prisoners to participate
in programs vary greatly, as do the circumstances
when they may be required to enroll. Incentives to
encourage inmate participation also vary and may
include receiving wages, gaining privileges, accumulating “good time,” or receiving a sentence reduction.
Such incentives are critical to encouraging participation and perseverance. The results of a survey on such
requirements and incentives are described below.
The 2004 publication cited previously, Correctional
Education, identifies the detailed information necessary for understanding the state of the prison education system. And there is a need for similar studies
to be conducted regularly, given the indications that
investment is declining at a time when the prison
population is growing and that the number of former
prisoners returning to their communities also is growing.
The availability of information about our public education system has been steadily improving, and even
more data are required under NCLB. We believe there
is urgency in creating a parallel prison information
system on enrollments, achievement, advancement,
and the quality of the curriculum and the teachers.
Prisoners have significantly fallen behind, and the time
they spend behind bars will likely be the best opportunity for preparing them to return to society.

20

Requirements and Incentives
There are a variety of actions taken by states to increase inmate participation in education, beyond simply making it available. On the requirement side, there
has been a trend toward mandatory participation over
the past two decades or so, a trend that stems from the
adoption of mandatory education requirements by the
Federal Bureau of Prisons in 1981. At the time of the
last survey in 2002, 44 percent of the states had passed
such mandatory requirements.20 Ten of the states that
mandate programs also provide various incentives for
participation. This kind of forced education remains
the subject of debate. Some believe it is inappropriate
or ineffective in getting inmates to learn, while others
point to research that shows otherwise. In some states,
of course, all of the classrooms are filled to capacity
with volunteers.
While the remaining states do not have mandatory education provisions in their laws or policies, in
several states judges may impose such a requirement
at the time of sentencing. And four of these states have
incentives of varying strength to encourage achievement, such as earning “good time,” or requiring a GED
to receive a raise in a prison industry. Table 4 provides a
brief summary of the various approaches taken by states.
The Declining Investment
Captive Students, an ETS report published in early
1996, reported a decline in the resources available for
education and training in prisons, as well as a wide
variation of resources among the states. According to
the report, at least half of all state correctional institutions had cut their inmate educational programs over
the prior five years.
Average state expenditures can be deceiving, as the
state-by-state budgets revealed. In 1993-94, the latest
time for which data were available, the total budget
per inmate varied from just under $2,500 in Minnesota
and about $1,300 in Vermont, down to almost nothing
in California, Mississippi, and Wisconsin. A middle
amount was about $500 in Arkansas. Similar large
variations occurred in the amount spent per partici-

For more information on trends and considerations involved in mandatory education programs, see Jerry McGlone, Status of Mandatory
Education in State Correctional Institutions, U.S. Department of Education, Office of Vocational and Adult Education, Sept. 20, 2002; and
Harold D. Jenkins, Mandatory Education: A Status Report, Nov. 22, 2002.

16

Table 4
Examples of Mandatory and Voluntary Education in State Correctional Institutions
Examples of Mandatory Education Requirements
Arizona

attainment of 8th-grade literacy tied to earned release credit

Florida

“gain time” for GED

Hawaii

education a pre-condition for parole

Montana

three-month commitment signed; non-completion has parole consequences

Oklahoma

GED influences parole decision

Pennsylvania

no job assignment for refusal to participate in education

Texas

program required for parole

Delaware

“good time” earned for voluntary participation

Indiana

adult basic education and GED programs are voluntary

Kentucky

pay and “good time” for participation

Minnesota

diploma or GED tied to raises

New Jersey

parole may be delayed for non-participation

Examples of Voluntary Education Programs

Source: Jerry McGlone, Status of Mandatory Education in State Correctional Institutions, U.S. Department of Education, Office of Vocational and
Adult Education, September 20, 2002, Figure 1.

pant — from more than $6,000 in New Jersey down to
just a few dollars in Alaska and Wisconsin. Figure 11,
reproduced from the Captive Students report, illustrates these differences.
The decline has continued. From 1990 to 2000, the
proportion of prison staff providing education fell
from 4.1 to 3.2 percent of the total staff. That, combined with the large increase in the inmate population,
boosted the number of inmates per instructor during
that period from 65.6 to 95.4, or 45 percent.21
An extreme example of decline in education investment
is Oregon. According to Gary Harkins, who started
working for the Oregon Department of Corrections
in 1980, an inmate then could learn a vocation or
study all the way to a Ph.D. These days, he says the
2,000-inmate Oregon State Penitentiary has not one
teacher on its staff.22
While the federal government has provided support
for state correctional programs since the mid-1960s,
declines are also evident:

• Before 1998, the federal government required states
to spend no less than 10 percent of their Basic State
Grant for Adult Education in state institutions,
including correctional institutions; the law now
requires them to spend no more than 10 percent.
• Under the Carl D. Perkins Vocational and Technical
Act, funds are provided to states for programs in
correctional institutions. Prior to 1998, states were
required to spend at least 1 percent of their grants
in state institutions (including correctional), but
the 1998 amendments specify that no more than 1
percent may be so spent. As a result, some states
have cut back on their expenditures in correctional
programs.
• As part of the “get tough on crime” doctrine, Pell
Grants that fund postsecondary education for lowincome students were, in 1994, denied to inmates.
• Correctional institutions have had difficulty qualifying for federal aid due to changes to the Library
Services and Construction Act in 1996.23

Klein et al., 2004, p. 20.
Peter Slevin, “Prison Experts See Opportunity for Improvement,” Washington Post, July 26, 2005.
23
Michelle Tolbert, State Correctional Education Programs: State Policy Update, National Institute for Literacy, March 2002.
21
22

17

Figure 11
Total Education Budget per Participant, 1993-94
While we have seen a long-term declining investment, some are optimistic about a turnaround. For
example, Marc Mauer, assistant director of the Sentencing Project based in Washington, D.C., says the
climate “has changed substantially,” adding, “There’s
a growing liberal-conservative consensus that it’s in
everyone’s interest that we provide resources in prison
that decrease the chances of recidivism.”24

New Jersey
Vermont
South Dakota
Minnesota
Connecticut
Nebraska
New Mexico
Virginia
Oregon
Tennessee
Florida
Illinois
Michigan
Maryland
Georgia
Montana
New York
Delaware
Arkansas
South Carolina
North Carolina
Colorado
Kansas
Iowa
Pennsylvania
New Hampshire
Hawaii
Mississippi
Rhode Island
Ohio
California
Kentucky
Massachusetts
Idaho
Alaska
Wisconsin
0

1,000

2,000

3,000 4,000

5,000

6,000 7,000

Dollars
Source: Corrections Compendium, Volume XIX, No. 3., March 1994. Missouri, Oklahoma,
and Wyoming are not included because their reported budgets do not include personnel
costs. Missing states did not respond to the survey.

24

Slevin, 2005

18

What Prison Education and Training Programs Accomplish

The use of education and training in prison programs
became pervasive in the 1930s. Since then, it has fluctuated with society’s alternating emphasis on rehabilitation and punishment. Despite this long history, careful studies of the effects of these efforts were slow in
coming. There have now been a considerable number
of studies and evidence of success is accumulating.

• Of 14 findings regarding recidivism, nine show positive effects.

A shift away from rehabilitation through education began in the 1970s. In 1975, Linton, Martinson,
and Wilks published an influential and widely known
assessment of efforts at rehabilitation.25 Their work
called into question the efficacy of most attempts at
rehabilitation, after a stretch of renewed optimism
and activism beginning in the 1960s. Martinson also
published a review of studies in 1974, with a similar
conclusion: Nothing works.

Vocational Education. The conclusion: “Most of
the research conducted in recent years shows a correlation between vocational training and a variety of
outcomes generally considered positive for society or
for correctional institutions.”

Lipton and colleagues did concede that “offenders
are amenable to training and education … (and) can
generally improve basic educational skills given the
teacher’s real concern, personal interest, and dynamic
instruction.” Missing, however, were hard data.
Almost 20 years later, Gerber and Fritsch completed
a comprehensive evaluation and summation of the
prior two decades’ research. The two also took another
look at the Lipton, Martinson, and Wilks review, and
disagreed with the conclusions drawn.26
Gerber and Fritsch’s review was an ambitious
undertaking. The two evaluated each study’s methodology, with ratings based on factors such as control
groups, matching vs. random assignments of subjects,
use of statistical controls, and use of tests of statistical
significance. Excluding the publications that met none
of the criteria for inclusion, they report on the results
of 72 studies, most of them conducted in the 1980s
and early 1990s. A brief summary of their conclusions
follows.
Basic Secondary Education. The conclusion: “A
few researchers found no evidence that adult academic
education has any positive effects on recidivism, but
the most common finding… is that inmates exposed to
education programs have lower recidivism rates than
nonparticipants.”

• Of four findings regarding post-release employment,
three show positive effects.
• Of two findings regarding post-release participation
in education, both show positive effects.

• Of 13 findings regarding recidivism, 10 show positive effects.
• Of seven findings regarding post-release employment, five show positive effects.
• Of two findings regarding disciplinary problems,
both show positive effects.
College Education. The conclusion: “Numerous
studies have shown a clear and fairly consistent correlation between collegiate studies and recidivism,
and between college and variables measuring personal
growth. At the same time, some critics have pointed
out methodological weaknesses in the research and
caution against overoptimistic interpretations.”
• Of 14 findings regarding recidivism, 10 show positive effects.
• Of three findings regarding post-release employment, all show positive effects.
• Of three findings regarding disciplinary problems,
one shows a positive effect.
• Of two findings dealing with post-release participation in education, both show positive effects.
Gerber and Fritsch identify factors that explain
why some programs are more successful than others
in achieving their stated goals. The researchers
draw upon reviews of 10 successful programs by Rice
et al.,27 Luiden and Perry,28 and themselves to
determine the following:

D. Lipton, R. Martinson, and J. Wilks, The Effectiveness of Correctional Treatment, New York: Praeger, 1975.
J. Gerber and Eric J. Fritsch, Prison Education and Offender Behavior: A Review of the Scientific Literature, Prison Education Research
Project, Report 1, July 1993.
27
E. Rice et al., Assessment of Quality Vocational Education in State Prisons, Executive Summary, Final Report, 1980.
28
R. Luiden and D. Perry, “An Evaluation of a Prison Education Program,” Canadian Journal of Criminology, 1984.
25
26

19

• The more extensive the educational program, the
more likely it is to succeed. In New York, for example, inmates who earned a GED were less likely
to return to prison than those who attended classes
but did not earn a GED.
• Programs that are separate from the rest of the
prison are more likely to succeed.
• Programs that follow up with inmates after their
release are more likely to succeed.
• Programs tailored to the prison population are
more likely to succeed.
• Programs that hone skills needed in the job market
are more likely to succeed.
Although the Gerber and Fritsch work did not address the need for programs of substantial duration,
rather than short-term classes, a subsequent examina-

Figure 12
Percentage Returning to Prison by the
Duration of Academic Programs

Percentage Returning to Prison

30

tion of 14,000 inmates released from Texas prisons in
1991 and 1992 suggests that duration is important (see
Figure 12). Adams et al. conducted the Texas study,
with help from Gerber and Fritsch.29
The Federal Bureau of Prisons in its Post-Release
Employment Project conducted a longitudinal study
of more than 7,000 inmates from 1983 to 1987. Its purpose was to follow up on federal prisoners to see what
effect the vocational or apprenticeship training and
prison work experience had after they were released
from prison. Prisoners who gained work experience in
prison industries were 24 percent less likely to recidivate, and those who participated in apprenticeship and
vocational training were 33 percent less likely. Lower
rates of reincarceration were found as many as eight
to 12 years after release.30
More recently, two researchers took a large step
toward pinning down the effect of prison education
and training on recidivism — and thus crime. Stephen
S. Steurer and Linda G. Smith published the ThreeState Recidivism Study in 2003. The uniqueness of
their effort — which involves the states of re-arrest,
re-conviction, and re-incarceration — is described by
Steurer in the preface:
The extensive exit survey given to all inmate participants before release… has not been done in other
research studies. This survey yielded data about the
offender’s family, prior involvement in the criminal
justice system both as a juvenile and an adult, educational attainment, employment, and release plans
which have never been collected from such a large
sample of offenders leaving prison until now. Finally,
no study has ever been able to collect and assemble
data from so many important sources — offenders
themselves, correctional, institutional and educational
records, parole officers, state and national criminal
history repositories, and state wage and labor data.31

25

20

15

10
None

100 or less

101-200

201-300

301 or more

Hours of Academic Programs
Source: Adams et al., 1994.

K. Adams et al., “A Large-Scale Multidimensional Test of the Effects of Prison Education Programs on Offenders’ Behavior,” The Prison
Journal, Vol. 74, No. 4, December 1994.
30
See W. Saylor and G. Gates, Correctional Management Quarterly, 1(2), 1997, and summarized in State Educational Programs, National
Institute for Literacy, March 2002, p. 20.
31
Stephen J. Steurer and Linda G. Smith, Education Reduces Crime: Three-State Recidivism Study, Centerville, UT: Management and
Training Corporation, 2003.
29

20

The new study strongly suggests that prison education and training can lead to increased employment
and significant reduction of recidivism — which, in
turn, means lower crime rates and costs associated
with building and staffing prisons.32 Overall results
from the three states can be seen in Figure 13. All of
the differences are statistically significant at the .01
level. The overall drop in recidivism rates is 29 percent. In the combined results, there were no significant
differences between participants and nonparticipants
in the nature of new offenses committed.

Figure 13
Impact of Education and Training on Recidivism
70

60

Non-participants

48

The most recent — and most rigorous — metaanalysis of the results of interventions, drawing on all
the latest research available, were those of Gaes
et al., in 1999 and of Aos et al., in 2001. The results are
summarized in the book Prisoners Once Removed, by
Jeremy Travis and Michelle Waul:
These recent meta-analyses continue to show
treatment effectiveness, particularly for academic
instruction, vocational training, cognitive skills,
sex offender programs, and substance abuse
intervention.34
These authors also point out that there are methodological flaws in some of the studies available.
Regardless, there is substantial reason to expect that
programs such as those studied can reduce recidivism
and prison costs.

57

Participants
50

Audrey Bazos and Jessica Hausman took this study
one step further. Using the three-state study, they did
a cost-benefit study and concluded that “one million
dollars spent on correctional education prevents 350
crimes” and is almost twice as cost-effective as the money
being spent on expanding the capacity of prisons.33

Percentage

40
35
31

30

27
21

20

10

0

Re-arrest

Re-conviction

Re-incarceration

Source: Steurer and Smith, 2003.

This study is not based on randomization — the gold standard in program evaluation — but the authors do attempt to measure motivation
and to control for it as best as they can.
33
Bazos and Hausman, 2004, p. 2.
34
Gerald G. Gaes and Newton Kendig, Prisoners Once Removed, Jeremy Travis and Michelle Waul (eds.), Washington, DC: Urban Institute,
2004, pp. 113-117. The authors also summarize a comprehensive cost-benefit analysis of interventions.
32

21

Prisoner Re-Entry Programs: A Recent Response

In the past several years, the problems of re-integrating
an annual flow of around 650,000 prisoners into communities has raised interest in rehabilitation to a level
not seen in the United States since back in the 1960s.
Elsewhere in this report, the retreat from a rehabilitation approach is described. The number of prisons and
the costs involved have become so huge that they are
daunting, even to those who prefer the strict-punishment
approach and the long-term removal of criminals from
society. The impact of 97 percent of this immense
population returning — at some point — to society
has broken into public and official consciousness.
A significant initial response came from the federal
government in 2001. The Department of Justice made
more than $100 million in grants available to states for
developing and expanding programs aimed at easing
re-entry after release.
The Council of State Governments furthered efforts
by establishing the politically bipartisan Re-Entry
Policy Council. Made up of 100 key leaders in communities and state, local, and federal governments —
including state lawmakers, criminal justice policymakers
and practitioners — the council includes the following:
• workforce development and employment services
officials
• housing providers and housing system officials
• representatives of health, mental health, and
substance abuse treatment systems
• victim advocates
• people who have been incarcerated and their
families
• ministers and others working in faith-based
institutions
Upon its formation, the Council spoke strongly
about what has been happening:
Despite their proven cost-effectiveness, prison
and jail-based services are already threadbare.
In nearly half the states, departments of correc-

tion are or have been under some form of federal
court supervision because of overcrowding or the
insufficiency of services available to inmates…
As even less emphasis is placed on the services
and supports people need upon their release from
prison and jail, extraordinary investments are
made in providing emergency services to people
whose condition has deteriorated to the point
that they cycle repeatedly through jails, emergency rooms, and detox facilities.”35
The report describes in detail, in its nearly 700 pages,
all of the elements necessary in re-entry programs,
and the kinds of partnerships required in the community. A model program, the Transition from Prison to
the Community Initiative (TPCI) is referred to and is
described in a 37-page companion piece issued by the
National Institute of Corrections.
Regarding its release, a New York Times editorial
commented, “State and federal lawmakers are finally
realizing that controlling prison costs means controlling recidivism — by helping newly released people
establish viable lives once they get out of jail.”36
Re-entry projects around the country have been
similar, or at least somewhat similar, to the Offender
Re-entry Program (ORP) in Suffolk County, Massachusetts. The lead educational agency is Boston’s Bunker
Hill Community College. The program is detailed in a
case study found in a report of the Economic Policy
Institute (EPI).37
ORP provides soon-to-be-released inmates with an
intensive 6-hour-a-day course of study over a six-week
period. After release, the inmates continue to receive
support from caseworkers and mentors for a minimum of six months — and many choose to continue
beyond this six-month period. Drawing on different
community resources and agencies, the program includes:
• education during the final six weeks of prison
provided by Bunker Hill Community College

Re-Entry Policy Council, Report of the Re-Entry Policy Council: Charting the Safe and Successful Return of Prisoners to the Community,
Executive Summary, Jan. 21, 2005, www.reentrypolicy.org/executive-summary.html.
36
New York Times, Jan. 21, 2005.
37
Anne M. Piehl, et al., Prospects for Prison Re-Entry, Working Paper No. 125, Economic Policy Institute, August 2003, pp. 6-8.
35

22

• job assistance at the one-stop career center
called Workplace
• case management provided by Community
Resources for Justice
• mentoring support from the faith-based
Ella J. Baker House
Texas Project RIO (for “Reintegration of Offenders”), an earlier program closely linked to the prison
system, provides incarcerated prisoners with a variety of services aimed at matching released prisoners
with jobs. Services include a week-long job search
workshop, job-placement assistance, and post-release
follow-up. A 1992 one-year follow-up reported that 69
percent of RIO participants found jobs, compared with
36 percent of non-RIO parolees, and that 23 percent
compared with 38 percent returned to prison.38 Initial
returns after an additional eight-month follow-up were
also encouraging, but results from longer-term followup studies are needed.
The prison-return problem has continued to gain
visibility. There is certainly a need for transition services and partnerships. But what about the need for
education, training, and useful work experience over
the duration of incarceration? A six-week program
just before release may be helpful and an important
complement to prior efforts, but it is much too short
term to raise educational achievement to the high
school level, or to impart a job skill demanded by the
labor market.
The idea of imparting the education and skills
people need to succeed out of prison is emerging from
the dark, but it has not yet seen the light of day.

38

Jeremy Travis, But They All Come Back: Facing the Challenges of Prisoner Reentry, Washington, DC: Urban Institute Press, 2005, p. 173.

23

The Children of Prisoners

When we think of prisons and prisoners, we think
of people who have broken the law. This image does
not include the children of prisoners. It is likely that
the great majority of these children had difficult lives
before a parent’s incarceration, given the typical low
educational levels and prior income of prisoners, and
the other kinds of disruptions that may have existed
in the children’s family life. Having a father convicted
and sent to prison exacerbates any pre-existing problems and jeopardizes the children’s well-being. And if a
child’s mother is convicted, the degree of disruption is
compounded since often the children of prisoners live
in mother-only families.
Separation from a parent affects children in many
ways, not the least of which is financially. In 1997,
seven in 10 parents in state prisons reported that they
were employed either full- or part-time just prior to incarceration. Sixty percent of fathers reported having a
full-time job (55 percent of the men are fathers), compared to 39 percent of mothers. And welfare to parents
under the Temporary Assistance for Needy Families
programs ceases upon incarceration (although the new
caregiver may receive payments for the children).39
To ignore what happens to the children of prisoners in the home, the school, and the community is to
accept that a high percentage of these children will
follow in their parents’ footsteps. In 2000, almost 3.6
million parents were either in prison or on parole,
and about 1.5 million children had parents who were
incarcerated in state or federal prisons. This number
is even higher when parents in jail are included. The
average age of children with at least one incarcerated
parent was eight.40
Not much is known about how the children of
prisoners are faring. What is known is that schools,
communities, and prison systems are doing little on an
organized basis to help them. Some of the efforts are
described in a report by Ross D. Parke and K. Alison
Clarke-Stewart.41

The California State Library published a review of
the situation for children of prisoners in California in
2000.42 At that time, an estimated 856,000 children in
California — or 9 percent of the state’s children — had
at least one parent who was involved in the adult
criminal justice system. Of those, 292,000 were estimated to have a parent in state prison or a county jail;
the rest had a parent on parole or probation.
A 1992 report by the California Assembly Office of
Research found little information on the children of
prisoners and reported that “these children are not
recognized as a group by any state agency or department in California.”43 According to the more recent
California State Library report, this is still the case,
with neither police nor the courts regularly inquiring
— at arrest or sentencing — whether a prisoner has
children.
The California State Library report summarizes
what little is known. A survey of mothers jailed in
Riverside found that over half of their children were
between 3 and 6 years old. A number of small-scale
studies found that:
“The children may suffer from multiple psychological problems, including trauma, anxiety, guilt,
shame, and fear. Negative behavioral manifestation can include sadness, withdrawal, low selfesteem, decline in school performance, truancy,
and use of drugs or alcohol and aggression.”44
Contact with an incarcerated parent is hard to
maintain in California, as in many places. While 60
percent of the state’s female prisoners are from Southern California, the two largest women’s prisons are
260 miles away. One survey of prisoners’ contact with
children concluded that, “Today’s prison system is
designed to discourage rather than encourage parent/
child relationships,” due largely to distance, visitation
restrictions, and associated costs.45

Travis, 2005, pp. 126-127.
Marian Wright Edelman’s Child Watch weblog, Aug. 22, 2003, www.childrensdefense.org/childwatch/030822.aspx.
41
Ross D. Parke and K. Alison Clarke-Stewart, Effects of Parental Incarceration on Young Children, paper produced for a conference funded
by the U.S. Department of Health and Human Services, December 2001, p. 3.
42
Charlene Wear Simmons, Children of Incarcerated Parents, California State Library, California Research Bureau, March 2000.
43
Sharron Lawhorn, Children of Incarcerated Parents: A Report to the Legislature Pursuant to ACR 38, Assembly Office of Research, May 1992,
cover letter, as cited in Simons, 2000, p. 3.
44
Simmons, 2000, p. 4.
45
James Austin, Patricia Hardyman, and John Irwin, “Exploring the Needs and Risks of the Returning Prisoner Population,” a paper
produced for a conference by the U.S. Department of Health and Human Services, Jan. 30-31, 2002, p. 73.
39
40

24

According to the California State Library Report,
an American Bar Association study found: “While law
enforcement policies and procedures specifically addressing children of arrestees may not currently exist
in most agencies, the issue of accountability — and
subsequent legal liability — is nevertheless present.”
Further, courts have asserted that police officers are
duty bound to assure the safety of children present at
an arrest. This short-term safeguard does not, however, assign accountability to any person or entity over
the duration of the parent’s incarceration.
Examples of long-term efforts to help children and
families do exist. The descriptions provided in the paper by Parke and Clarke-Stewart illustrate what is possible, and provide a basis for thinking about broader
policy. Current efforts are carried out by prison socialwork agencies, schools, and clinics:
• Parenting education for incarcerated mothers.
Studies have shown positive results in the kinds of
knowledge needed. There are examples also of efforts with fathers.
• The family unit as a target for intervention.
While most efforts have focused on the incarcerated parent, some are directed at the family unit,
stemming “from claims that post-release success is
higher among inmates who have maintained family ties during incarceration.” This may include
conjugal visits, furloughs, and family and marital
counseling. The United States lags behind other
countries in taking this approach; Mexico, Sweden,
Denmark, and Canada are cited.
• Visitation programs. Some women’s institutions
have visiting programs that provide play areas for
children, extended visits, and even special housing
in the institution for children. Crafts and games
may be provided, as well as transportation.
• “Co-detention:” Raising children in prison. There
are innovative programs in Europe and the United
States that allow for mother and child to remain together for some period of time. Nurseries in prison
go back to early in the 20th century in the United
States. A strong argument is that such arrange-

46

ments permit the mother to develop an emotional
attachment with the child. Drawbacks include the
appropriateness of the environment for children
and the degree of freedom had by the children.
• Alternatives to incarceration. Some type of community-based sentencing is involved. This might
include house arrest; halfway houses; or day programs at correctional institutions with the mother
returning home at night. One survey of 24 community-based programs reported reduced recidivism
and increased family preservation.
An example of efforts to increase contact between
prisoners and their families is Hope House in Washington, D.C. According to Jeremy Travis:
Hope House connects incarcerated fathers with
their children in the District, hosts summer
camps at federal prisons in North Carolina and
Maryland where children spend several hours a
day for a week visiting with their fathers in prison… (and) created a teleconference hookup with
federal prisons in North Carolina, Ohio, and
New Mexico so that children can go to a neighborhood site to talk to their fathers in prison.46
The ever-increasing rate of incarceration of
mothers and fathers makes the question of the
children unavoidable. Ignoring this question instead
makes unavoidable a new generation destined for
the fate of its parents. The magnitude of the problem
is sobering. If an average of 9 percent of California
children are affected by a parent’s contact with the
criminal justice system, what is the percentage among
minority and economically disadvantaged children?
The tide has been turning for decades toward
punitive measures, and the downward trend in prison
education and training has been a steep one. No matter where people stand on whether sentences should
be longer or the three-strikes-and-you’re-out approach,
no one can deny the problem of ignoring the children
and what it may mean — to them and for a safe society — in the future.

Travis, 2005, p. 137.

25

On the contrary, incarceration has to be looked at in
its broader context, as pointed out by Jeremy Travis:
Imprisonment causes ripple effects that are felt
throughout a prisoner’s family network and has
magnified those effects in a strong undercurrent that is eroding the familial infrastructure of
America’s poorest communities. Virtually every
social institution that deals with children —
including families, schools, child welfare
agencies, foster care, and kinship care systems
— is touched by the high rates of imprisonment.47
Today, interest in protecting the parent-child bond
in relation to prisoners is on the rise and is extending
to fathers. The Vera Institute of Justice conducted “A
Review of the Field,” concluding that:
Despite the absence of formal public policies and
minimal public recognition of need, parenting
programs are offered in a few prisons and jails,
though they have not had anywhere near the
longevity experienced by programs designed for
women’s prisons.48
There is no reason why the disagreements over
crime and punishment should extend to caring for
prisoners’ children. Everyone has reason to be concerned with whether the children of convicted criminals grow and prosper, and become law-abiding and
productive citizens. What must still be resolved is
whether new policies or approaches are sound and
effective.

47
48

Travis, 2005, p. 147.
John M. Jeffries et al., Serving Incarcerated and Ex-Offender Fathers and Their Families: A Review of the Field, Vera Institute of Justice, 2001,
p. 8.

26

The Special Case of the Black Male

Table 5
Black Male Incarceration Rates, by Age Groups
Ages
18-19

Ages
20-24

Ages
25-29

Ages
30-34

Ages
35-39

Ages
40-44

Ages
45-54

Ages
55 Up

5.4%

11.1%

12.6%

11.0%

10.0%

8.0%

4.6%

0.9%

Source: Calculated from data in Harrison and Beck, 2005.

In a general picture of high and rising incarceration
rates, the situation of Black males is divergent enough
to require an in-depth look. As seen in Figure 4, the
incarceration rate for Black males is seven times that
of White males and three times higher than Hispanic
males. These multiples are typical at each age interval,
except 35 to 39, when the Black male rate drops to 3.4
times the White male rate.
Overall, the incarceration rates for Black males
are staggering, as shown in Table 5. The low rate of
incarceration at age 55 and over reminds us that,
unless they die in prison, all prisoners return to society
at some time.
To put the incarceration rates of Black men into
perspective, consider the following: When the national
unemployment rate rises to 10 percent or more, we
characterize the economy as past a recession and in
a depression. If at least 10 percent of U.S. men in this
age range were fighting a war, the country would
experience serious challenges to its productivity. And
if that percentage were hit by a deadly virus, the
proportion would be labeled epidemic.
These incarceration rates are part of a larger
phenomenon in the Black community and cannot
be viewed in isolation. Black males, particularly in
their 20s and 30s, are disappearing from the economy
— and from traditional family units. The two are, of
course, connected. About half of these young men are
leaving high school without a diploma. These are the
men who are in the most serious straits.
Unemployment rates for high school dropouts are
staggering. But these rates only reflect those who are

49
50

looking for work; many more dropouts give up and
become classified as “not in the labor force.”
Even after combining categories, the numbers do
not provide a full picture. Disengagement and alienation extend much further. The statistics represented
above can only reflect the “civilian non-institutional
population” — those whom the U.S. Census Bureau
finds living in households when the Bureau does its
household canvasses and labor force surveys. But substantial numbers of Black males are not even counted;
they are not found in the household when the censustaker knocks at the door. While extensive efforts by
the Census Bureau to reduce this “undercount” have
met with considerable success, the undercount has by
no means been eliminated. For example, from 1960 to
2000, the estimated percent of uncounted 18- to 29year-old Black males dropped from 15.1 percent to 6.5
or 8.1 percent, depending on the demographic analysis
model used.49 Regardless of the model, that’s still a lot
of men missing.
And then there are those who are accurately counted because they are in prison. The reason for examining more than the incarceration rate is that, for a large
proportion of Black men, there is a movement among
these categories: being unemployed, being out of the
labor force, being disengaged to the extent of not being found by census takers, and being locked up. From
prison, the odds are high of returning to one of the
earlier categories — and then repeating the cycle.
Figure 14 represents a slice of the whole as represented by Black males with less than a full high school
education and by those with a high school education,
for 1970 and 2000.50

J. Gregory Robinson, Accuracy and Coverage Evaluation: Demographic Analysis Results, U.S. Bureau of the Census, Table 7, March 2001.
Steven Raphael, The Socio-Economic Status of Black Males: The Increasing Importance of Incarceration, Goldman School of Public Policy,
University of California, Berkeley, March 2004, Tables 2 and 3.

27

Figure 14

If we specifically focus on dropouts ages 18 to 25
and 26 to 30, we see an even bleaker picture. Among
18- to 25-year-old dropouts in 1970, 50 percent were
employed; in 2000, just 27 percent were. In 1970, 8
percent were institutionalized; in 2000, 23 percent
were — almost as much as the percent who were
employed.

Trends in the Status of Black Males,
by Education Level, 1970 and 2000

High School Dropouts
Employed

71

34

Among 26- to 30-year-olds, 76 percent were
employed in 1970, with only 30 percent employed in
2000. Six percent were institutionalized in 1970, with
that number rising to 34 percent in 2000 — surpassing
the percentage who were employed. High school graduates at these ages fared better. However, even among
graduates, the percent of 18- to 24-year-olds employed
dropped from 62 percent in 1970 to 44 percent in
2000; the employment rate for 26- to 30-year-olds fell
from 83 percent to 58 percent over the same time
period.

Not working, not looking
23

47

In armed forces

1
<1

Institutionalized
4

19

1970
2000

High School Graduates
Employed
56

75

Not working, not looking
15

35

In armed forces
2

8

Institutionalized
2

0

Over 60 percent indicated that they would “probably not” or “definitely not” hire applicants with
criminal history records, with “probably not” being the
modal response.

8
20

Prisoners returning to their communities, particularly without at least a high school education, have
three strikes against them in getting a job. First, a
criminal record makes it difficult to secure employment upon re-entry to society. Even when employment
is secured, the amount of earnings is adversely affected. Second is a lack of work experience, since time
when others would be bolstering their credentials was
spent in prison. Third, some occupations are closed
to felons, including jobs requiring close contact with
children, various health and public service occupations, and jobs in security firms.51 Employer surveys
reveal an aversion to hiring ex-prisoners; Raphael cites
the Multi-City Study of Urban Inequality:

40

60
Percentage

80

100

Employers surveyed were those who had recently
hired low-skilled workers — which ex-prisoners are
likely to be.52

Source: Raphael, 2004.

51
52

Raphael, 2004, p. 21.
For an in-depth analysis of the effect of incarceration on subsequent labor market success, see Henry Holzer et al., Declining Employment
Among Young Black Less-Educated Men: The Role of Incarceration and Child Support, National Poverty Center Working Paper #04-5, April
2004.

28

Ex-offenders, of whatever race or ethnicity, are
at the bottom of employers’ wish list. Surveys show
that employers would much prefer to hire someone
on welfare, someone whose highest level of academic
achievement is a GED, someone with a spotty work
history, or someone who has been unemployed for a
year — all people who normally have a relatively hard
time in the labor market.53
While we know how many people are in prison at a
given point in time, we have less reliable information
regarding how many members of society have a prison
record, irrespective of when they served time. Such
data would better depict the extent to which ex-prisoners face employment problems.
The Bureau of Justice Statistics regularly makes
estimates by race and age, but not by education level.
It estimates that 20 percent of 24- to 35-year-old Black
males and that 22 percent of 35- to 44-year-old Black
males have served time. The corresponding percentages for White males are 3 and 4; for Hispanic males,
they are 9 and 10. Another Justice Policy Institute
study estimated that by 1999, 50 percent of Black male
high school dropouts in their early 30s had prison
records, compared to 10 percent of White males.

How these dropouts fare depends largely on their family supports and pressures.
Given the trends here described, the prospects for
increasing equality in employment and earnings, and
in the achievement of children in school, seem dim.
Dropouts that do end up in prison should not be
dismissed as lost causes. Instead, their time incarcerated should be spent raising their educational level,
receiving realistic training for jobs, and earning quality work experience through prison industries. The
obstacles of being an ex-prisoner are real enough; raising prisoners’ education and skill levels high enough
to be competitive is the best way we can help them
successfully transition back to the workforce, society,
and their families.

“These findings clearly show that, for low-education
African American men, prison has become a common
life event, even more common than employment or
military service,” states Princeton University’s Bruce
Western, a co-author of the JPI report.54
The situation in California is worse than that in the
nation as a whole. Raphael used administrative records
to make estimates for California. He found that almost
all Black males who were dropouts had served time.55
The labor market prospects for the large proportion of young Black males who do not complete high
school are dire and have been steadily worsening, as
they have for dropouts generally. The prospects for
those with a prison record are considerably worse.

Travis, 2005, p. 164.
Justice Policy Institute release, Aug. 28, 2003, downloaded June 17, 2005, regarding a policy brief, Education and Incarceration.
55
Raphael, 2004, Table 4. Raphael explains the methodological problems in making these estimates for California; they probably should be
taken as approximations (see p. 22 and the footnotes to Table 4).
53
54

29

Conclusion

Those who are comforted by the flow of criminals into
prison tend to forget that those who go in will likely
come out — whether or not they are prepared to return
to society and function as productive citizens.
The prison population largely includes people
with the lowest levels of educational achievement.
Those who go to prison often have not done well in
the employment world either. The 600,000 or so exoffenders returning to the community each year are
in danger of being as locked out of the labor market
as they were locked up in jail, particularly if they have
not completed high school. They re-enter society with
anywhere from one to three of the following strikes
against them:
• Strike One — Ex-prisoners, many of whom are
high school dropouts, have an increasingly hard
time securing a job; if they get one, they have difficulty earning a wage on which they can live and
support their families.
• Strike Two — Employers value experience and continuity of work history. Former prisoners have been
out of the labor market and have not added to their
work experience.
• Strike Three — Employers are reluctant to hire
former prisoners.
The rate of incarceration for Black male dropouts
between 20 and 39 has reached a critical point, with
an almost equal number imprisoned as employed. The
likelihood of raising their achievement levels, and their
children’s in turn, is poor — unless they gain education while they are in prison.
In the 1970s, the United States began rejecting
efforts directed at rehabilitation — efforts such as education and training. Many had concluded that rehabilitation did not work.
The analysis behind that view has since been proven
wrong, and many subsequent studies have found that
education and training in prison can reduce recidivism

30

and help ex-offenders readjust to community life.
Recently, some prisoner re-entry programs — typically
of short duration, held just before a prisoner is released,
and that provide continuing transition services — have
shown positive results.
Despite a consistently growing need to prepare
prisoners for life outside bars, as well as to protect
citizens from harm and to reduce costs associated
with incarceration, the investment in prison education has fallen, decade by decade. Reversing that trend
is a minimum requirement, not a wholesale solution,
for reducing the high rates of recidivism. Education,
training, and transition support at the point of release
are imperative but should not be viewed as a panacea
— or rejected because they are not.
Another dimension of the prisoner problem has
gained recognition: Prisoners are parents of some 1.5
million children, children who are often neglected and
who live in circumstances that put them at high risk
of following in their parents’ footsteps. To reduce the
risk of these children turning to a life of crime, their
parents must be able to return to society as selfsupporting, responsible adults.
In the United States, as the late Senator Daniel
Patrick Moynihan said, we only do what we measure.
That we are not measuring the prison education
enterprise in ways that would permit making decisions to improve it is a strong statement of the low
priority assigned to prisoner education. Having made
notable strides in measuring the effectiveness of public
schools, the departments of Education and Justice
now need to make a comparable effort to measure the
quantity and effectiveness of education in prisons.
There is some new hope for extending and improving education for juvenile offenders in the mandates
of NCLB so that delinquent youth also benefit from
“best education” practices. The National Collaboration
Project has shown that many programs for juvenile
offenders will need a lot of help fulfilling the NCLB

mandates. Nineteen states have reported that they are
unable to show whether they are meeting the requirements for “adequate yearly progress.”56
The trend toward punishment and toughness need
not serve as an impediment to education and training
efforts. Learning is tough work, particularly for those
who have not been encouraged to treat it as a priority
or who have developed an aversion to it. But a minimum of a high school education is required for self
and family support, and the criminal justice system
has a responsibility to best serve society. What better
way to do this than to ensure the self-sufficiency of
ex-prisoners and to help break the cycle of crime in
prisoners’ families?
A chain gang requires hard physical labor; a learning gang requires hard mental effort and discipline.

56
57

A growing number of states are understanding this
and are enacting requirements and incentives to increase the educational attainment of prisoners. While
approaches are still debated, there are precedents and
experience on which to build.
We end this report with an excerpt from Captive
Students: Education and Training in America’s Prisons,
a report we issued a decade ago.
We are polarized as a nation on the question of
how to deal with crime and how to treat prisoners. Perhaps we are much less polarized on
the question of whether it is in our self-interest
to make sure our ex-prisoners are literate. This
is the question we raise by issuing this report:
should these captives also be students?57

John Gehring, “NCLB’s Mandates on Delinquent Youth Get Attention,” Education Week, July 27, 2005.
Barton and Coley, 1996, p. 31.

31

32

Listening. Learning. Leading
0864 ITEM #